GAMETE TRAITS AND MEASURES OF ADULT BODY CONDITION
Female egg size was assessed from three replicate counts using a Beckman multisizer™ 3 Coulter counter (mean number of eggs per female = 8,371 ± 1140 SE). To estimate sperm size, digital photographs of sperm were taken using a digital camera attached to a microscope (OLYMPUS IX53, 20X objective). Sperm length was measured for each male using the software ImageJ (http://rsb.info.nih.gov/ij). Sperm length was measured in μm as the distance from the tip of the head (excluding the protruding acrosomal cap or ‘beak’) to the end of the tail. Ten sperm were measured for each male, and the average of the 10 measures was used in the analysis.
Sperm motility was assessed using computer-assisted sperm analysis (CEROS sperm tracker, Hamilton-Thorne, Beverley USA). Sperm motility was assessed at two sperm concentrations (106 and 108). For each individual, two 4 μl aliquots of the sperm solution were placed in two separate wells of a 12-well Multi-test slide and then covered with a coverslip. Sperm motility was therefore assessed in two subsamples for each male at both sperm concentrations. Given the high reported repeatability of sperm motility measures in blue mussels (Fitzpatrick et al. 2012), we recorded the average of the two subsamples for the ensuing analysis, and caution was taken to analyse sperm in a random order with respect to the sperm concentration used. From the CASA analyses, we obtained the percentage of motile sperm and a series of parameters describing velocity and trajectory of motile sperm. These parameters include average path velocity (VAP), straight-line velocity (VSL), curvilinear velocity (VCL), lateral head displacement (ALH), beat cross frequency (BCF), straightness (STR) and linearity (LIN). Cut offs for static cells were the same as those used in previous experiments by Eads et al. (2016). An average of 276.8 ± 37.0 SE sperm tracks were analysed for each male at each concentration.
As body condition may influence fertilisation success (i.e. mussels in better condition may have higher quality gametes) we controlled for this potential confounding effect in our analysis. We recorded shell length (mm) and flesh mass (g) from each brood parent and used these to calculate a condition index using the residuals of flesh weight against shell length.